toneden-automation

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.65). The required runtime workflow calls Rube MCP tools like RUBE_SEARCH_TOOLS/RUBE_MANAGE_CONNECTIONS, which can return tool schemas and execution plans as readable text from an external MCP server (outsider-authored content) that the agent ingests into its LLM context.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill requires adding and using the MCP endpoint https://rube.app/mcp at runtime to fetch tool schemas, recommended execution plans, and connection state, which directly control the agent's prompts/execution flow, so this external URL is a required runtime dependency that can dictate agent behavior.